Segment structure of UoE and UOE

[Shubham A]

I tried analysing the acronym ‘UoE’ and it’s variation in Festival. I get a different pronunciation and hence, segment structure for ‘UoE’ and ‘UOE’. Here are the commands:

>(set! utt1 (SayText “UoE!”))
>(utt.relation.print utt1 ‘Segment)
()
id _10 ; name pau ; dur_factor 0 ; end 0.22 ; source_end 0.106461 ;
id _6 ; name y ; dur_factor 0.102492 ; end 0.275619 ; source_end 0.284943 ;
id _7 ; name uw ; dur_factor 0.123242 ; end 0.399283 ; source_end 0.403568 ;
id _9 ; name ow ; dur_factor 0.7486 ; end 0.578798 ; source_end 0.630109 ;
id _11 ; name pau ; dur_factor 2 ; end 1.0276 ; source_end 0.972081 ;
nil

>(set! utt2 (SayText “UOE!”))
>(utt.relation.print utt2 ‘Segment)
()
id _14 ; name pau ; dur_factor 0 ; end 0.22 ; source_end 0.106461 ;
id _8 ; name y ; dur_factor 0.736476 ; end 0.293053 ; source_end 0.284943 ;
id _9 ; name uw ; dur_factor 1.00642 ; end 0.459464 ; source_end 0.403568 ;
id _11 ; name ow ; dur_factor 0.785204 ; end 0.640549 ; source_end 0.554692 ;
id _13 ; name iy ; dur_factor 2.30826 ; end 0.851352 ; source_end 0.823731 ;
id _15 ; name pau ; dur_factor 2 ; end 1.30015 ; source_end 1.22565 ;
nil

However, the segments are same for ‘CSTR’ and ‘CsTR’. Any explanation?

[Simon]

You need to investigate the earlier steps in the pipeline to see where the difference between “UoE” and “UOE” first arises.

[Author]

Posts